Abstract: There are provided systems and methods for identifying data processing operations on websites using network traffic logs. A service provider server may provide website and application tools that may track user navigations, inputs and selections, and other operations with merchant websites and/or online merchant marketplaces. Further, the service provider may provide discounts and offers during purchases and transactions with the online merchants, which may further provide reward points or value based on use of the service provider's services. In order to determine available payment options on merchant websites and/or via merchant applications, the service provider may process network traffic logs and other network traffic data. This may include scripts and codes to identify assets for payment service providers. The service provider may utilize an inference mode of a scripted parsing operation to identify these assets and payment options, and record with domains in order to log such availability.

Abstract: Techniques are disclosed relating to item representations. A computer system may access information identifying a first set of items and generate a representation of the first set of items that positions them in an embedding space. The computer system may send a request to another computer system for information pertaining to an item selected from the first set of items and receive correlation information that identifies recorded user behavior indicative of correlations between a second set of items and the selected item. The computer system may update the representation based on the correlation information, such that at least one of the first set of items that is included in the second set is moved closer to the selected item in the embedding space and at least one of the first set of items that is not included in the second set is moved farther away from the selected item.

Abstract: Methods and systems are presented for stolen cookie detection. An authentication request is received for a user to access a website using a web browser executable at the user's device. A series of storage locations available on the device for storing web cookies is identified and sorted in order of increasing fraud risk starting from a first storage location. A cookie value for each storage location is retrieved from the device. For each storage location after the first: an expected cookie value is calculated based on the cookie value of a preceding storage location; the expected cookie value is compared with the value retrieved for the storage location; and a score representing a level of fraud risk for the storage location is assigned. The authentication request is processed based on whether the assigned score for at least one of the storage locations exceeds a predetermined risk tolerance for fraud detection.

Abstract: Methods and systems are presented for providing explainable artificial intelligence for a deep-learning model on a global level. Multiple surrogate models are generated based on characteristics of the deep-learning model, where each surrogate model is configured to mimic a behavior of the deep-learning model with respect to one of the output dimensions associated with the deep-learning model. Simulations are performed on the surrogate models. Based on the simulation results, importance scores are calculated for each input feature of the deep-learning model. An importance score calculated for an input feature represents an extent to which the input feature contributes to a corresponding one of the output dimensions associated with the deep-learning model. The importance scores may then be used to modify the characteristics of the deep-learning model or other downstream machine learning models.

Abstract: Methods and systems are presented for automatically detecting positions of various webpage elements within a webpage when the webpage is rendered, based on analyzing the programming code of the webpage. A position detection system obtains and parses the programming code of the webpage to identify webpage elements within the webpage. A group of related webpage elements is identified based on a shared programming structure. The position detection system generates a DOM tree based on the programming code, and determines relative positions of the webpage elements within the group by traversing the DOM tree using a breadth-first search algorithm.

Abstract: Aspects of the present disclosure involve systems, methods, devices, and the like for classifying or indexing websites using an efficient multistage statistical indexing scheme. In one embodiment, a system is introduced that includes a plurality of models for classification. The classification and indexing may be obtained using at least two models designed to retrieve information from a website visited for marketing and assessing a user. In one embodiment, a first model, content-based features are used for the classification and in making predictions of the unlabeled observations. In another embodiment, a second model includes embedded technology-based features for further classification.

Abstract: Systems and methods for hierarchical segmentation and classification may include receiving an input object having a hierarchical data structure. The input object may be segmented at each hierarchical level in a descending manner. The components of each hierarchical level may be classified in an ascending manner. Classification at higher levels may depend on the results from classification at the lower levels. After classification, a hierarchically-classified object may be outputted.

Abstract: Methods and systems are presented for automatically detecting positions of various webpage elements within a webpage when the webpage is rendered, based on analyzing the programming code of the webpage using graph-based and NLP-based techniques. A position detection system obtains the programming code of the webpage and generates a DOM tree based on the programming code. A group of leaf tags may be identified based on traversing the DOM tree. The position detection system may use two different, but coordinated, prediction models to make the classification prediction for the leaf tags, where the first prediction model focuses on recall by identifying as many positive leaf tags as possible at the cost of low precision and the second prediction model refines the results from the first prediction model by improving the precision of the classification.

Abstract: A computer system identifies that a first portion of markup language, extracted from a markup language document of a website, corresponds to a first actionable element, wherein the first portion of markup language is a variable length representation. In response to identifying that the first portion of markup language corresponds to the first actionable element, the computer system utilizes a recurrent neural network (RNN) encoder to create a first code representation that corresponds to the first portion of markup language. The computer system identifies a first additional information that corresponds to one or more pre-defined goals. The computer system creates a final fixed length markup language representation that includes the first code representation and the first additional information. The computer system inputs the final fixed length markup language representation into a model.

Abstract: Methods and systems are presented for automatically detecting positions of various webpage elements within a webpage when the webpage is rendered, based on analyzing the programming code of the webpage. A position detection system obtains and parses the programming code of the webpage to identify webpage elements within the webpage. A group of related webpage elements is identified based on a shared programming structure. The position detection system generates a DOM tree based on the programming code, and determines relative positions of the webpage elements within the group by traversing the DOM tree using a breadth-first search algorithm.

Abstract: Methods and systems are presented for automatically detecting positions of various webpage elements within a webpage when the webpage is rendered, based on analyzing the programming code of the webpage. A position detection system obtains and parses the programming code of the webpage to identify webpage elements within the webpage. A group of related webpage elements is identified based on a shared programming structure. The position detection system generates a DOM tree based on the programming code, and determines relative positions of the webpage elements within the group by traversing the DOM tree using a breadth-first search algorithm.

Abstract: Methods and systems are presented for automatically detecting positions of various webpage elements within a webpage when the webpage is rendered, based on analyzing the programming code of the webpage using graph-based and NLP-based techniques. A position detection system obtains the programming code of the webpage and generates a DOM tree based on the programming code. A group of leaf tags may be identified based on traversing the DOM tree. The position detection system may use two different, but coordinated, prediction models to make the classification prediction for the leaf tags, where the first prediction model focuses on recall by identifying as many positive leaf tags as possible at the cost of low precision and the second prediction model refines the results from the first prediction model by improving the precision of the classification.

Abstract: Systems and methods for hierarchical segmentation and classification may include receiving an input object having a hierarchical data structure. The input object may be segmented at each hierarchical level in a descending manner. The components of each hierarchical level may be classified in an ascending manner. Classification at higher levels may depend on the results from classification at the lower levels. After classification, a hierarchically-classified object may be outputted.

Abstract: A computer system identifies that a first portion of markup language, extracted from a markup language document of a website, corresponds to a first actionable element, wherein the first portion of markup language is a variable length representation. In response to identifying that the first portion of markup language corresponds to the first actionable element, the computer system utilizes a recurrent neural network (RNN) encoder to create a first code representation that corresponds to the first portion of markup language. The computer system identifies a first additional information that corresponds to one or more pre-defined goals. The computer system creates a final fixed length markup language representation that includes the first code representation and the first additional information. The computer system inputs the final fixed length markup language representation into a model.

Abstract: Aspects of the present disclosure involve systems, methods, devices, and the like for classifying or indexing websites using an efficient multistage statistical indexing scheme. In one embodiment, a system is introduced that includes a plurality of models for classification. The classification and indexing may be obtained using at least two models designed to retrieve information from a website visited for marketing and assessing a user. In one embodiment, a first model, content-based features are used for the classification and in making predictions of the unlabeled observations. In another embodiment, a second model includes embedded technology-based features for further classification.

Abstract: Methods and systems are presented for automatically detecting positions of various webpage elements within a webpage when the webpage is rendered, based on analyzing the programming code of the webpage. A position detection system obtains and parses the programming code of the webpage to identify webpage elements within the webpage. A group of related webpage elements is identified based on a shared programming structure. The position detection system generates a DOM tree based on the programming code, and determines relative positions of the webpage elements within the group by traversing the DOM tree using a breadth-first search algorithm.

Abstract: Aspects of the present disclosure involve systems, methods, devices, and the like for classifying or indexing websites using an efficient multistage statistical indexing scheme. In one embodiment, a system is introduced that includes a plurality of models for classification. The classification and indexing may be obtained using at least two models designed to retrieve information from a website visited for marketing and assessing a user. In one embodiment, a first model, content-based features are used for the classification and in making predictions of the unlabeled observations. In another embodiment, a second model includes embedded technology-based features for further classification.

Abstract: A computer system identifies that a first portion of markup language, extracted from a markup language document of a website, corresponds to a first actionable element, wherein the first portion of markup language is a variable length representation. In response to identifying that the first portion of markup language corresponds to the first actionable element, the computer system utilizes a recurrent neural network (RNN) encoder to create a first code representation that corresponds to the first portion of markup language. The computer system identifies a first additional information that corresponds to one or more pre-defined goals. The computer system creates a final fixed length markup language representation that includes the first code representation and the first additional information. The computer system inputs the final fixed length markup language representation into a model.

Abstract: A computer system identifies that a first portion of markup language, extracted from a markup language document of a website, corresponds to a first actionable element, wherein the first portion of markup language is a variable length representation. In response to identifying that the first portion of markup language corresponds to the first actionable element, the computer system utilizes a recurrent neural network (RNN) encoder to create a first code representation that corresponds to the first portion of markup language. The computer system identifies a first additional information that corresponds to one or more pre-defined goals. The computer system creates a final fixed length markup language representation that includes the first code representation and the first additional information. The computer system inputs the final fixed length markup language representation into a model.

Abstract: Aspects of the present disclosure involve systems, methods, devices, and the like for classifying or indexing websites using an efficient multistage statistical indexing scheme. In one embodiment, a system is introduced that includes a plurality of models for classification. The classification and indexing may be obtained using at least two models designed to retrieve information from a website visited for marketing and assessing a user. In one embodiment, a first model, content-based features are used for the classification and in making predictions of the unlabeled observations. In another embodiment, a second model includes embedded technology-based features for further classification.